Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Conducting multiple linear regression via sequential analysis

esvesv
edited July 2016 in JASP & BayesFactor

Hi all,

Before I begin, I would like to state that my statistical knowledge is pretty limited, so I apologize in advance if my queries sound elementary.

The goals of my current research project require me to 'statistically control' for covariates (e.g., fluid intelligence) so that I can evaluate the unique contributions of the variables of interest to the model. Traditionally, in SPSS, I would do this by first adding the covariates to Step 1 and subsequently including the variables of interest to Step 2 in a multiple linear regression. However, I am trying to see if I can do this using a Bayesian approach in JASP.

In JASP, I noticed that when I perform a Bayesian linear regression, there is a drop-down tab entitled 'Model' where I can select certain variables as 'nuisance' - does this mean I'm designating them as covariates? If true, does this achieve the same goal as 'statistically controlling' for covariates akin to a sequential analysis in multiple linear regression?

Any assistance on this matter would be very much appreciated. Thank you!

Comments

  • Hi ESV,

    Yes, when you design them as nuisance they get added as covariates. This is indeed similar to a hierarchical analyses where you add batches of covariates in a particular order.

    Cheers,
    E.J.

Sign In or Register to comment.